In view of the issues of missed and false detections encountered in small object detection for UAV remote sensing images, and the inadequacy of existing algorithms in terms of complexity and generalization ability, we propose a small object detection model named IA-YOLOv8 in this paper. This model integrates the intra-group multi-scale fusion attention mechanism and the adaptive weighted feature fusion approach. In the feature extraction phase, the model employs a hybrid pooling strategy that combines Avg and Max pooling to replace the single Max pooling operation used in the original SPPF framework. Such modifications enhance the model’s ability to capture the minute features of small objects. In addition, an adaptive feature fusion module is introduced, which is capable of automatically adjusting the weights based on the significance and contribution of features at different scales to improve the detection sensitivity for small objects. Simultaneously, a lightweight intra-group multi-scale fusion attention module is implemented, which aims to effectively mitigate background interference and enhance the saliency of small objects. Experimental results indicate that the proposed IA-YOLOv8 model has a parameter quantity of 10.9 MB, attaining an average precision (mAP) value of 42.1% on the Visdrone2019 test set, an mAP value of 82.3% on the DIOR test set, and an mAP value of 39.8% on the AI-TOD test set. All these results outperform the existing detection algorithms, demonstrating the superior performance of the IA-YOLOv8 model in the task of small object detection for UAV remote sensing.

Small-object detection in UAV remote sensing faces common challenges such as tiny target size, blurred features, and severe background interference. Furthermore, single imaging modalities exhibit limited representation capability in complex environments. To address these issues, this paper proposes CTU-YOLO, a UAV-based small-object detection algorithm built upon cross-modal feature fusion and lightweight upsampling. The algorithm incorporates a dynamic and adaptive cross-modal feature fusion (DCFF) module, which achieves efficient feature alignment and fusion by combining frequency-domain analysis with convolutional operations. Additionally, a lightweight upsampling module (LUS) is introduced, integrating dynamic sampling and depthwise separable convolution to enhance the recovery of fine details for small objects. Experiments on the DroneVehicle and LLVIP datasets demonstrate that CTU-YOLO achieves 73.9% mAP on DroneVehicle and 96.9% AP on LLVIP, outperforming existing mainstream methods. Meanwhile, the model possesses only 4.2 MB parameters and 13.8 GFLOPs computational cost, with inference speeds reaching 129.9 FPS on DroneVehicle and 135.1 FPS on LLVIP. This exhibits an excellent lightweight design and real-time performance while maintaining high accuracy. Ablation studies confirm that both the DCFF and LUS modules contribute significantly to performance gains. Visualization analysis further indicates that the proposed method can accurately preserve the structure of small objects even under nighttime, low-light, and multi-scale background conditions, demonstrating strong robustness.

Small target detection in Unmanned Aerial Vehicle (UAV) remote sensing images faces challenges such as insufficient feature information and background interference in complex environments. Existing single-modal methods, which rely solely on one data source, struggle to address these issues effectively. This paper proposes CA-YOLO, a lightweight detection framework based on adaptive cross-modal feature fusion and an attention mechanism. The framework employs a dual-branch backbone network to extract complementary features from RGB and IR modalities, utilizes an Adaptive Cross-modal Feature Fusion (ACFF) module to dynamically integrate global semantics and local details, and leverages a Lightweight Hybrid Attention (LHA) mechanism to enhance targets and suppress noise. Experiments on three public datasets, DroneVehicle, LLVIP, and FLIR, demonstrate that CA-YOLO achieves mAPs of 75.1%, 96.1%, and 71.4%, respectively, while requiring only 6.8 MB parameters and 16.4 GFLOPs computational cost, significantly outperforming state-of-the-art methods. The inference speeds reach 57.1 FPS, 78.1 FPS, and 56.5 FPS on the respective datasets, achieving an optimal balance between accuracy and efficiency. Ablation studies and visualization results further validate the effectiveness of each module and the framework's superior capability in reducing missed detection and false alarm rates.

Recently, hyperspectral image (HSI) mixed denoising methods based on nonlocal subspace representation (NSR) have achieved significant success. However, most of these methods focus on optimizing the denoiser for representation coefficient images (RCIs) without considering how to construct RCIs that better inherit the spatial structure of the clean HSI, thereby affecting subsequent denoising performance. Although existing works have constructed RCIs from the perspective of sparse principal component analysis (SPCA), the refinement of RCIs in mixed noise conditions still leaves much to be desired. To address the aforementioned challenges, in this paper, we reconstructed robust RCIs based on SPCA in mixed noise circumstances to better preserve the spatial structure of the clean HSI. Furthermore, we propose to utilize the robust RCIs as prior information and perform iterative denoising in the denoiser that incorporates low-rank approximation. Extensive experiments conducted on both simulated and real HSI datasets demonstrate that the proposed robust RCIs guidance and low-rank approximation method, denoted as RRGNLA, exhibits competitive performance in terms of mixed denoising accuracy and computational efficiency. For instance, on the Washington DC Mall (WDC) dataset in Case 3, the denoising quantitative metrics of the mean peak signal-to-noise ratio (MPSNR), mean structural similarity index (MSSIM), and spectral angle mean (SAM) are 36.06 dB, 0.963, and 3.449, respectively, with a running time of 35.24 s. On the Pavia University (PaU) dataset in Case 4, the denoising quantitative metrics of MPSNR, MSSIM, and SAM are 34.34 dB, 0.924, and 5.505, respectively, with a running time of 32.79 s.

This paper presents a lightweight hyper-network architecture for cross-scale multi-task object classification, addressing the critical challenge of gradient interference in joint learning scenarios. We propose a HyperConv module integrated into a slim ResNet-12 backbone, which dynamically generates task-adaptive 3 × 3 convolutional kernels from compact two-dimensional latent vectors. This design allows explicit control over gradient flows for different tasks with minimal parameter overhead (only 3.2% additional parameters). Our framework incorporates adversarial regularization via a Gradient Reversal Layer (GRL) and dynamic task-weight scheduling to mitigate gradient conflicts across domains. Experiments on both natural image datasets (Mini-ImageNet and CIFAR-100) and remote sensing benchmarks (EuroSat and UCMerced_LandUse) demonstrate statistically significant improvements over conventional shared-parameter baselines. The proposed method effectively reduces negative transfer, enhances feature representation, and offers a practical solution for on-device multi-task learning in resource-constrained remote sensing applications such as UAVs and edge satellites.